This project applies stimulus control technology to fundamental problems in teaching individuals with severe learning difficulties to relate words and referents to graphic symbols for use in selection-based communication (SBC) systems. The overall goal is to develop effective, efficient methods for assessing and teaching relations among such stimuli, leading to the development of stimulus equivalence classes. We view these efforts as supporting ongoing applied research in SBC training for difficult-to-teach individuals. Positive outcomes will have theoretical implications as well: They will provide the first demonstration of stimulus equivalence in frankly nonverbal humans. Our study sample comprises students with severe learning difficulties who lack both functional vocal speech and the fine motor and imitation skills necessary to learn manual signing, but who could learn to communicate by selecting graphic symbols to indicate choices, needs, comments, and questions. Intervention studies have shown that SBC systems (i.e., communication books, boards, and electronic devices) can function as effective communication output modes as well as teaching tools for many individuals with mental retardation and autism. Those with the most severe learning difficulties, however, often do not respond to efforts to teach them by conventional methods that symbols and words can substitute for referents. Project 6 takes a relational learning approach to this critical problem. Project 6 will take advantage of technological developments to study several procedures for teaching relations among symbols, referents, and spoken words. Our studies use laboratory-validated stimulus control procedures such as match-to-sample (MTS) with exclusion and stimulus control shaping to teach relations among stimuli that constitute functional SBC vocabulary items. These procedures are combined with technological innovations to enable rich computer simulations of rudimentary SBC training. Techniques to be explored here include computerized presentations of: (1) high-quality motion video, animation, and scanned photographs; (2) recorded human speech. We will also evaluate transfer from structured training situations to use of symbols for recorded human speech. We will also evaluate transfer from structured training situations to use of symbols for communication in less structured, everyday situations. Specific Aims of Project 6 are to: (1) assess the discriminability of sets of stimuli that are commonly used in SBC systems; (2) analyze the entering symbol matching repertoires of our samples of individuals with severe learning difficulties; (3) investigate methods for teaching relations between symbols and referents that minimize the probability of errors occurring in training, and result in functional substitutability of symbols for referents. (4) examine the function of synthetic speech output in the initial development of spoken word/symbol/referent relations; (5) explore techniques for developing new spoken word/symbol relations rapidly; (6) apply stimulus equivalence analyses to the expansion of stimulus equivalence classes for users of SBC systems; and (7) examine the relationship between language comprehension and acquisition of rudimentary SBC skills.